git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH v4] send-email: make it easy to discern the messages for each patch

From
DSDragan Simic <dsimic@manjaro.org>
Date
Apr 8, 2024, 16:15 UTC
Message-ID
<d334e0cc3a681d7ec5b06edac7d21232@manjaro.org>
In-Reply-To
<xmqqttkbetd3.fsf@gitster.g>
Hello Junio,
On 2024-04-08 18:01, Junio C Hamano wrote:
Show 22 quoted lines
> Dragan Simic <dsimic@manjaro.org> writes:
> 
>> On 2024-04-06 19:05, Junio C Hamano wrote:
>>> Junio C Hamano <gitster@pobox.com> writes:
>>> 
>>>> The whole thing deserves to be a three-patch series, the first one
>>>> being a preliminarly "let's move the final newline out of the
>>>> translatable string" step, followed by "let's have a gap between
>>>> output for each patch sent out".  Perhaps another "even during
>>>> sending a single patch, we may want extra blank lines when use of
>>>> editor and other user interation is involved" patch on top.
>>> Or the latter two could be done in a single patch.  I'll have to
>>> re-review the thing (if I were the only reviewer of the topic) so
>>> doing so would delay the completion of the topic, though.
>> 
>> Huh, I've already separated this patch into three patches, and IMHO
>> they look nice and make everything less error prone.  Would you agree
>> with the three-patch approach, please?
> 
> My "or the latter two could be done in a single patch" was
> "alternatively you can", so either way is fine as long as the result
> is well structured.

Great, thanks. I really appreciate your highly detailed approach to reviewing the patches.

Show 9 quoted lines
> I know how to explain "insert a gap between
> patches" well.  I do not know which one is easier to explain,
> between
> 
>  (1) now we have "insert a gap between patches" with patch [2/3],
>      but when editor invocation and confirmation prompts are
>      involved, there are these three cases where we want to tweak
>      the logic to show gaps.  Here in patch [3/3] I explain how each
>      of these three affect the logic from the previous step.

In my opinion, the insertion of vertical whitespace can also be seen rather independently when it comes to separating the patches, and separating the prompts from the patches.

Show 7 quoted lines
>  (2) We want to insert a gap before showing the second and
>      subsequent patches, unless in such and such cases.  We also
>      want to insert a gap when we do this and that.  We do all of
>      these in this patch [2/2].
> 
> So doing it in three-patches results in a series easier to
> understand by readers, by all means, please do.

As I mentioned above, the rather independent nature of the insertion of the two "classes" of vertical whitespace makes the three-patch approach look more suitable to me. I think it's also a better option if some regression is discovered later, because the patches can be applied and tested separately.

I already went ahead and submitted the three-patch series, [1] please have a look. I hope you'll like it.

[1] https://lore.kernel.org/git/cover.1712486910.git.dsimic@manjaro.org/T/#u

Previous: Junio C Hamano
Message 8 of 8 in “send-email: make it easy to discern the messages for each patch”
  1. send-email: make it easy to discern the messages for each patchDragan Simic, Apr 6, 2024
  2. Junio C HamanoApr 6, 2024
  3. Dragan SimicApr 6, 2024
  4. Dragan SimicApr 6, 2024
  5. Junio C HamanoApr 6, 2024
  6. Dragan SimicApr 6, 2024
  7. Junio C HamanoApr 8, 2024
  8. Dragan SimicApr 8, 2024

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.